[GH] (texera/multi-region-e2e-test): Workflow run "Benchmarks" failed!

[GitBox <gi...@apache.org>]

The GitHub Actions job "Benchmarks" on texera.git/multi-region-e2e-test has failed.
Run started by GitHub user aglinxinyuan (triggered by aglinxinyuan).

Head commit for run:
6616dadb4d185f722f3a81efff32f1bb22ea7b4b / Xinyuan Lin <li...@gmail.com>
test(amber): add X-shaped multi-region workflow e2e integration test

Adds MultiRegionWorkflowIntegrationSpec, an @IntegrationTest that runs a big
X-shaped workflow end-to-end across multiple regions: two Python sources fan
into a hash join (build/probe) feeding a Python sort UDF, and also into a
union; the sort and union are the two terminal (materialized) outputs.

Cross-region data delivery goes through input-port materialization reader
threads (which replaced cache-source operators in #3425). Until now that path
had no end-to-end coverage -- only scheduling unit tests counted regions on a
physical plan. This spec guards it before the dead cache-source plumbing is
removed (#6126).

The test is a black box: it builds a logical plan and runs it through the real
compiler, controller, and scheduler, then asserts only on the materialized
outputs -- the self-join on the unique key yields 100 rows returned sorted by
the Python UDF, and the union concatenates both sources into 200 rows. Correct
results are only possible if the cross-region delivery worked; the region count
itself stays covered by the scheduling unit tests.

Python-tagged, so it runs in the amber-integration CI job.

Report URL: https://github.com/apache/texera/actions/runs/28731396704

With regards,
GitHub Actions via GitBox